Adaptive Image Digital Watermarking Algorithm Based On Best Scramble

With the rapid development of computer network and multimedia technology, people need an effective method for copyright protection of digital multimedia more and more. As one of the important applications of information hiding technology in computer field, digital watermarking provides a new method to protect the copyright of multimedia information, and ensures the multimedia products are used legally. It embeds watermarks into digital multimedia, and carries out the functions such as information hiding, secret storing, copyright protection by detecting the watermarks that can track the use of digital multimedia.This paper takes still images as example, and introduces basic theory and character of image watermarking. The typical algorithm and performance evaluation methods of watermarking are analyzed. The wavelet transform theory and its application in watermarking are also explained. This paper researches on some important topics such as the selection of watermark information, preprocess methods of watermark, choosing of wavelet bases and decomposing level, positions for embedding watermark and taking advantage of human visual system(HVS). To solve the contradiction between robustness and imperceptibility of watermarking, an adaptive digital watermarking algorithm based on best scramble is presented. In this paper the main research findings and innovation are described as following:(1) Research on preprocess of watermark signal in detail. The distances between original pixels and scrambling pixels are always defined as scramble degree in current watermarking system. Only the positions of pixels are calculated in this definition, and the pixel values are ignored. So there is a serious limit in this definition. This paper imports image local variance as scramble degree into watermarking system, and presents a best shuffling method based on it, which provides new criterion measure on preprocess effect of watermark signal.(2) Research on Embedding strategy and algorithm in wavelet domain. Taking advantage of the features of human visual system(HVS), with the sensitivity of human eye towards local brightness and texture activity, the embedding strength can adjust adaptively to solve the contradiction between robustness and imperceptibility of watermarking. This paper also presents a new simple method to measure the texture activity based on wavelet tree structure, and this method is more effective.(3) According to the adaptive watermarking algorithm based on best scramble, an integrated image digital watermarking system is developed, and a lot of attack experiments are performed. Experimental results show that when the image is partly broken, the error bits are dispersed as more as possible because of best scramble, which ensures watermark can be successfully detected. Because the behavior of human visual system(HVS) is taken into consideration when the watermark is embedded, the invisibility of watermark and quality of image are improved. The attack experimental results show that it is extremely robust against normal image processing and kinds of attacks.